Psychosomatic Medicine Faster Service from Outside North America
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Norman, S. A.
Right arrow Articles by Diamond, M. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Norman, S. A.
Right arrow Articles by Diamond, M. P.
Related Collections
Right arrow Pain
Right arrow Sexual Medicine: Female
Right arrow Stress and Coping
Right arrow Therapeutic Interventions
Psychosomatic Medicine 66:174-183 (2004)
© 2004 American Psychosomatic Society

ORIGINAL ARTICLES

For Whom Does It Work? Moderators of the Effects of Written Emotional Disclosure in a Randomized Trial Among Women With Chronic Pelvic Pain

Sally A. Norman, PhD, Mark A. Lumley, PhD, John A. Dooley, PhD and Michael P. Diamond, MD

From the Departments of Psychology (S.A.N., M.A.L.), Anesthesiology (J.A.D.), and Obstetrics and Gynecology (M.P.D.), Wayne State University, Detroit, MI.

Address correspondence and reprint requests to Mark A. Lumley, PhD, Department of Psychology, Wayne State University, 71 West Warren Avenue, Detroit, MI. E-mail: mlumley{at}sun.science.wayne.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSION
 ACKNOWLEDGMENTS
 REFERENCES
 
OBJECTIVE: Although written emotional disclosure has potential as a stress management intervention for people with health problems, the main (group) effects of disclosure in medical populations are limited. This study sought to identify individual difference moderators of the effects of written disclosure among women with chronic pelvic pain.

METHODS: In a prospective, randomized trial, 48 women with chronic pelvic pain completed 3 individual difference measures and then wrote for 3 days about stressful consequences of their pain (disclosure) or positive events (control). Health status was assessed at baseline and 2 months after writing.

RESULTS: Main effect group comparisons indicated that disclosure writing resulted in significantly lower evaluative pain intensity ratings than control writing at follow-up, but there were no main effects on other outcome variables (sensory or affective pain, disability, affect). Three baseline individual difference measures, however, significantly moderated group effects. Compared with control writing, disclosure led to less disability among women with higher baseline ambivalence over emotional expression or higher catastrophizing, and to increased positive affect among women with higher baseline negative affect. Ambivalence, but not catastrophizing, was independent of negative affect in its moderation effect.

CONCLUSION: Although the main effects of writing about the stress of pelvic pain are limited, women with higher baseline ambivalence about emotional expression or negative affect appear to respond more positively to this intervention.

Key Words: pelvic pain, • expressive writing, • disclosure, • negative affect, • ambivalence, • moderator.

Abbreviations: CPP = chronic pelvic pain;; PANAS-X = Positive and Negative Affect Schedule, Expanded Version;; MPQ = McGill Pain Questionnaire;; SIP = Sickness Impact Profile;; AEQ = Ambivalence Over Emotional Expression Questionnaire;; CSQ = Coping Strategies Questionnaire.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 CONCLUSION
 ACKNOWLEDGMENTS
 REFERENCES
 
The effects of writing about stressful experiences have been examined using the emotional disclosure or expressive writing paradigm (1). In the standard version of this paradigm, participants are randomized to a disclosure group, in which they write about stressful experiences over several days for approximately 20 minutes daily, or to a control group, in which they write about neutral topics for an equivalent duration. Group differences in health and functioning are examined several months after writing. Most studies of disclosure have been conducted on healthy samples (eg, students), and these studies are fairly consistent in showing that disclosure leads to a range of benefits (2–9).

Some studies have examined disclosure’s effect on people with health or emotional problems, but these studies provide less consistent support for the benefits of disclosure than do the studies of healthy people. A few of these studies found beneficial group main effects on key outcomes after disclosure (10–13). Other studies, however, reported positive effects for only 1 or a few outcomes, whereas many other outcomes in those studies did not reveal main effects (14–17). Other studies found no main effects of disclosure, but identified a subset of patients who benefited (18, 19). Still other published (20–25) and unpublished (26–28) studies reported no benefits after disclosure in various populations. Finally, at least 2 studies reported some negative effects of disclosure, such as increased symptoms (19, 29).

Moderators of the Effects of Written Emotional Disclosure
The available literature, therefore, suggests that when people with medical or emotional problems engage in disclosure, there is only a modest benefit. Furthermore, even when disclosure leads to significant health improvements, only a minority of disclosing participants respond positively (11). There are undoubtedly many reasons for these limited main effects (eg, insufficient statistical power, inadequate follow-up times, unreliability in procedures and measures). We have proposed that the limited effects are also a result of including people in disclosure studies who may not need disclosure, who may be reluctant to engage fully in it, or who may have difficulty with the cognitive and affective tasks needed to process and resolve negative emotional experiences (30). Thus, research is needed to determine moderators of the effects of disclosure—that is, individual difference variables that predict who is most or least likely to benefit. Note that individual differences can moderate the effects of disclosure even when main effects are absent, because main effects may be attenuated by the inclusion of people who either did not respond or even worsened.

This study examined 3 variables that research or theory suggests potentially moderate disclosure’s effects. Ambivalence over emotional expression, the first potential moderator, denotes the conscious desire to express feelings accompanied by a reluctance to do so (31). The theory underlying the disclosure paradigm posits that people who are aware of negative emotions but who avoid, inhibit, or suppress them are the most likely to benefit from disclosure (1). We hypothesize that ambivalent people will benefit more from disclosure than less ambivalent people, because the perceived safety of expressing oneself in writing, particularly to an audience that one need not face, should overcome the ambivalence. Some, but not all studies (32), support this. For example, participants who wrote about topics that they had previously not shared with others showed the greatest health improvements (7), and people who reported difficulty talking about their feelings had better mood after expressive writing (33).

Pain catastrophizing, the second potential moderator in this study, is defined as the tendency to focus on and exaggerate the threat value of painful stimuli and to evaluate negatively one’s ability to cope with pain (34). We hypothesize that people who tend to catastrophize will respond better to disclosure than those who do not because disclosure provides a mechanism for sharing and processing avoided emotions and because it is a technique that may facilitate the management of stress and, potentially, pain. One study examined the effects of briefly writing about dental fears among people awaiting dental hygiene treatment. Disclosure participants who were pain catastrophizers responded better during the treatment than control participants who were catastrophizers (35).

Negative affect, the third potential moderator, was included for 2 reasons. First, negative affect may itself moderate disclosure’s effects. Although some studies have found no moderating effect of baseline depression or anxiety (1, 36), other studies have shown that people who report low negative affect benefit less from disclosure (37, 38). Thus, additional research is needed to clarify whether negative affect moderates disclosure’s effects. The second reason for including negative affect is that this construct is widely recognized as a potential confound when studying the effects of other self-reported measures. Negative affect is viewed as a manifestation of a basic personality trait (neuroticism or negative affectivity) or even as a response bias. Therefore, other narrow or specific constructs should demonstrate their independence from negative affect (39). Thus, we tested whether ambivalence over emotional expression and catastrophizing moderated disclosure’s effects independently from negative affect.

Chronic Pelvic Pain
Chronic pelvic pain (CPP) is defined as noncyclical (occurring outside of menses) pain in the pelvic region lasting for at least 6 months (40). Approximately 9.2 million women in the United States between age 18 and 50 years suffer from CPP (41). Diagnosis is difficult because many medical conditions or tissues can cause or contribute to CPP, including gynecologic (eg, endometriosis), gastrointestinal (eg, irritable bowel syndrome), and urologic disorders (eg, interstitial cystitis), and musculoskeletal anomalies (eg, hip dysfunction, herniated discs; 42–44). Often the source of the pain is unclear, and invasive exploratory or surgical procedures are performed, potentially resulting in iatrogenic pain from postoperative surgical adhesions (eg, scar tissue).

Women with CPP experience not only pain but also many secondary difficulties, including physical disability, mood disturbance, interpersonal problems, and sexual dysfunction (41, 45, 46). Some women with CPP are infertile, and others worry about potential infertility as they consider a hysterectomy to obtain pain relief. Conflicts with health professionals are common, in part because providers often feel helpless in treating CPP, and they may invalidate patients by attributing CPP solely to emotional problems, particularly sexual problems or abuse (47). Moreover, unlike many pain problems, the private nature of CPP makes it difficult to discuss with others. This stress, shame, and secrecy likely exacerbate the pain, disability, and mood problems accompanying CPP, leading to a vicious stress and pain cycle.

Importance of Instructional Sets for Disclosure and Control Conditions
Some disclosure studies instruct participants to write about any stressful experience they choose, whereas other studies instruct participants to write about reactions that are secondary to a particular stressor, such as having breast cancer (13). There are several reasons that the present study had women with CPP write about stress secondary to CPP rather than general stress. First, writing about specific current problems appears to yield larger effect sizes than writing about general stress (48). Second, writing about the effects of CPP may be particularly appropriate, because CPP is private and typically not discussed openly, and the psychosocial consequences of CPP are often not shared with physicians. Third, having patients write about the stressful consequences of a condition likely will be more validating and lead to greater adherence than having patients write about possible premorbid stressors (eg, childhood abuse).

The best control condition to use in disclosure studies has been debated. Early studies had controls write about trivial topics, but this approach may lack face validity and risks creating negative responses. Writing about emotionally neutral topics, such as daily activities or time management, overcomes some concerns but does not control for writing about emotionally engaging topics. We thought that a stronger control would be to have participants write about positive experiences, which is emotionally relevant and has face validity to reduce stress because it is consistent with suggestions to "count one’s blessings" or think positively.

Goal and Hypotheses
This randomized, prospective study of women with CPP compared a group that wrote about the stressful consequences of CPP for 3 days with controls who wrote about positive life events unrelated to CPP. We hypothesized that mood changes immediately after writing would parallel the affective valence of the writing (48)—that is, that writing about stress would lead to a more immediate negative mood than positive writing. We also assessed changes in pain, disability, and affect 2 months after writing, and because pain is multidimensional, we differentiated sensory, affective, and evaluative aspects of pain. We tested for main effects between disclosure and control writing, but given the inconsistent findings in the disclosure literature for medical patients, we did not expect robust main effects. Rather, we were primarily interested in individual differences. Thus, we assessed ambivalence over emotional expression, catastrophizing, and negative affect at baseline, and tested these variables as moderators of the effects of disclosure and control writing. Finally, to test the independence of ambivalence and catastrophizing from negative affect, we retested whether ambivalence and catastrophizing remained significant moderators when negative affect was simultaneously considered.


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 CONCLUSION
 ACKNOWLEDGMENTS
 REFERENCES
 
Participants
Participants were women experiencing pelvic pain for at least the previous 6 months, for a minimum of 7 days per month, with at least 2 days unrelated to menses. Women with a history of a psychotic diagnosis, recent (last 2 months) pelvic or abdominal surgery, recent (last 2 weeks) initiation or change in hormone treatment, current pregnancy, or surgery scheduled during the course of the study were excluded. The 48 women in the final sample averaged 38.2 years of age (range, 18 to 64 years; SD, 11.5) and 14.9 years of education (range, 11 to 18 years; SD, 2.1); 83% were European American, and the rest were African American. The duration of CPP averaged 12.7 years (SD, 2.3), and the women had an average of 2.2 (SD, 2.7) pelvic or lower abdominal surgeries. Women in this sample reported experiencing pain for an average of 77.0% (SD, 27.2) of the time during a typical month. They reported a range of presumed causes of the CPP, particularly endometriosis (50%) and interstitial cystitis (12.5%), and 1 case each of pelvic inflammatory disease, complex regional pain syndrome, surgical adhesions, and musculoskeletal problems. Yet 1/3 of the women had no known or suspected cause of CPP. Finally, 75% of the women reported a history of depression, either situational (48%) or pervasive (27%), and 64.6% of the women reported that they had received mental health treatment at some point in their lives.

Procedures
Potential participants were recruited via brochures at gynecology and pain clinics, mass media advertisements, and announcements from the Endometriosis Association. Recruitment began in March 1999 and ended in June 2000, with the last follow-up assessment completed in August 2000. Interested women telephoned the investigator, who screened them for criteria. The study was described as "designed to get a better understanding of the experience of pelvic pain and to explore the effects of a writing exercise on pain, mood, and functioning." Women who remained interested met individually with the investigator and gave written consent to the Institutional Review Board-approved protocol.

Participants were interviewed for demographics and medical history and given the baseline health status and moderator measures to complete. (Because the baseline assessment was extensive, some participants did not have time to finish and were allowed to do so at home and promptly mail back the baseline measures.) Participants were also given a sealed packet, which contained the instructions for either the disclosure group or positive control group along with 3 writing journals (1 for each day), each consisting of prewriting and postwriting mood questionnaires, the post-only essay rating scale, and several blank sheets of paper for writing. These packets were prepared in advance, numbered with a unique identifier, and randomized in blocks of 2; the interviewer was blind to group assignment. These packets were distributed to participants in the order that they were seen, and participants took the sealed writing packets home along with a stamped, preaddressed envelope with instructions to return the writing packets at the end of the 3 writing days. Writing packets instructed all participants to write for at least 20 minutes each day, for 3 days in a row, in a private place. The researcher telephoned each participant at least once during the writing period to verify that writing was occurring or to encourage the participant to write. Immediately before and after each writing session, participants completed the mood rating questionnaire, and immediately after writing, they completed several ratings of that day’s essay (eg, its meaningfulness). Two months after completing the writing, participants were mailed the health status measures, along with a preaddressed, stamped envelope to return them. After measures were returned, participants were mailed a check for $25 for participating. We attempted to debrief all participants who did not immediately complete either the writings or follow-up health status measures to encourage participation or determine reasons for withdrawal.

Experimental Groups
Disclosure Group
The instructions to this group asked participants to write about negative emotional experiences related to CPP, and included the following:

"Write about how pelvic pain has affected your life, problems you have experienced because of your pain, and your feelings about those experiences. You may write about problems that you mentioned on the questionnaires that you completed, or other difficulties that pelvic pain has caused for you. I want you to write in detail about a specific problem the pelvic pain has caused for you, and your emotions related to the problem. Ideally, you should write about a problem that you have not shared with others, or that you have been reluctant to discuss. You can write about a different pain-related problem each day, or about the same one for all three days."

Positive Control Group
The instructions to this group asked them to write about positive emotional experiences unrelated to CPP and included the following:

"Write about positive aspects of your life. You may write about an area of your life that is positive and is not affected by your pelvic pain, a recent event that made you feel good emotionally, or a fond memory from your past. The key is to focus on pleasant thoughts and feelings when you write. Think about a positive event in your life and write about it. Write the details and facts surrounding the event, and about the good feelings that you experienced. Do not write about your pelvic pain. You can write about the same event on all three writing days, or a different one each day."

Measures: Manipulation Check
Participants’ Ratings of Their Essays
Immediately after writing each day, participants completed a rating form, using a 7-point scale (1 = not at all; 7 = a great deal) to rate that day’s writing along 2 dimensions: "how meaningful it was for you" (meaningfulness), and "how much you have held back from actively talking with others about the topic" (holding back).

Measures: Immediate Mood
Immediately before and after writing each day, participants rated adjectives taken from the Positive and Negative Affect Schedule-Expanded Version (PANAS-X; 49) regarding "how you feel right now" on a 7-point scale (1 = not at all; 7 = a great deal). For this study, we used the anger, guilt, sadness, fear, and happiness subscales.

Measures: Baseline and 2-Month Follow-up
Pain was measured with the McGill Pain Questionnaire (MPQ; 50), which assesses a person’s pain experience on 3 dimensions using numerous pain-related adjectives. The sensory-discriminative dimension has 10 adjective sets and assesses nociceptive (eg, temporal, spatial, pressure, thermal) pain properties. The affective-motivational dimension has 5 adjective sets and assesses the emotional and autonomic qualities of pain. The evaluative-cognitive dimension assesses the overall intensity of pain and is thought to involve higher cortical processes about the probability of outcomes. This last dimension is composed of a single set of 5 adjectives (0 = no word selected, 1 = annoying, 2 = troublesome, 3 = miserable, 4 = intense, and 5 = unbearable). Participants circled the word within each group that reflected their pelvic pain during the previous 2 weeks. For each of these 3 dimensions, the mean rating for the words endorsed in that dimension was calculated. This measure is reliable and has been widely used and validated (50), and the 3-dimensional structure of the scale has been supported (51).

Disability in various domains was assessed with 2 scales from the Sickness Impact Profile (SIP; 52): the physical functioning scale, which taps the respondent’s perceptions about problems with mobility, ambulation, basic body care, and movement; and the daily functioning scale, which assesses daily activities such as recreation, work, household maintenance, sleep, and eating. Participants checked those items that described their recent functioning with respect to their CPP; higher scores reflected more severe disability in that domain. The SIP has been widely used and validated (52) and reflects changes in response to treatment (53).

Positive affect and negative affect were assessed with the 2, 10-item scales from the PANAS-X (49). Respondents rated the frequency that they experienced each item during the previous 2 weeks on a 5-point scale (1 = not at all; 5 = extremely).

Measures: Potential Moderators
Ambivalence over emotional expression was assessed with the Ambivalence Over Emotional Expression Questionnaire (AEQ; 31). Participants rated each statement on a 5-point scale (1 = "I have never felt like this;" 5 = "I frequently feel like this"). Ratings were summed across items; higher scores indicated more ambivalence. The AEQ has good internal consistency and stability over a period of 6 weeks, predicts poorer health and emotional functioning, and is inversely related to emotional expressiveness (31). The internal consistency ({alpha}) in our sample was 0.87.

Catastrophizing was assessed with the 6-item subscale from the Coping Strategies Questionnaire (CSQ; 34). Items were rated from 0 ("never do that") to 6 ("always do that") and averaged. Higher means indicated greater pain catastrophizing. The CSQ is the most widely used coping instrument in the pain literature, and the catastrophizing scale has consistently been related to maladaptive outcomes (54). The coefficient {alpha} in our sample was 0.84.

The baseline value of the negative affect subscale from the PANAS-X was used not only as a baseline health outcome measure but also as a potential moderator measure. The {alpha} at baseline was 0.88.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 CONCLUSION
 ACKNOWLEDGMENTS
 REFERENCES
 
Participant Attrition and Group Comparisons at Baseline
A total of 78 women met inclusion criteria, were interested in participating, and provided consent. Of these, 18 (23.1%) dropped out without completing the baseline assessment. The most common explanations were that there were too many questionnaires or that some of them were too upsetting (particularly the SIP). Of the 60 women who completed baseline measures and were randomized, 10 (2 disclosure, 8 control) dropped out before finishing the writing task, and 2 (both disclosure) dropped out after writing but before follow-up. Reasons for dropping out were obtained from 8 of these 12 women: writing about stress was too upsetting (N = 2), it was depressing not being able to think of positive things to write about (N = 2), or the death of a family member (N = 1) or other life circumstances (N = 3) prevented them from finishing. Participants who dropped out after completing baseline measures (N = 12) were compared with completers (N = 48) on demographics, baseline values of the dependent measures, and group assignment using t tests or {chi}2. Dropouts were significantly younger, less educated, and had higher baseline negative affect and affective pain than completers. Dropouts and completers did not differ on other variables or assignment to experimental groups.

The completers in the disclosure (N = 28) and control (N = 20) groups were compared to determine the success of randomization. The groups did not differ on age, education, ethnic composition, duration of CPP, number of surgeries, or percent of the month with pain (all p >.17). The 2 groups also did not differ on the baseline levels of the any of the dependent measures, which are listed in the baseline columns of Table 1 (all p >.23).


View this table:
[in this window]
[in a new window]
  		TABLE 1. Baseline, Follow-up, and Change Scores (Mean, SD) for the Health Status Dependent Measures for the Disclosure and Control Groups (N = 48)
 
Adherence and Manipulation Checks
All 48 women returned their writings, and all but 1 (a disclosure subject who wrote for 1 day) wrote for all 3 days. We conducted checks on the success of the manipulation. First, a rater who was blind to group assignment read each woman’s essays and judged which group she was in. All but 1 person was correctly classified. Second, essays were reviewed for content. Disclosure participants wrote about many stressful consequences of CPP: negative mood effects (85.8% of the disclosure participants), activities impeded (60.7%), fear of an undetected medical condition (59.2%), feeling misunderstood (53.5%), frustration with the medical system (50.0%), feeling helpless (42.9%), negative impact of CPP on self-image (42.9%), impact of CPP on sexual functioning (39.3%), and jealousy of healthy people (35.7%). Control women also adhered to the instructions; 18 (90%) wrote only about pleasant events (eg, vacations, weddings, or times with their children), and the other 2 wrote about difficult events that were resolved with positive outcomes. Third, groups were compared on their holding back and meaningfulness ratings of their writings (averaged over the 3 days). The disclosure group’s rating of holding back (mean, 4.89; SD, 1.46) was significantly higher than the control group’s rating (mean, 2.28; SD, 1.48; t[46] = 6.06; p < .001). In contrast, the disclosure group’s rating of meaningfulness (mean, 5.62; SD, 1.29) was slightly lower than, but not significantly different than, the control group’s rating (mean, 6.18; SD, 0.98; t[45] = -1.64; p = .11). Thus, as we had hoped, both groups wrote about personally meaningful issues, but the disclosure group was more likely to have avoided talking about these issues.

Effects of Writing on Immediate Mood
We expected disclosure writing to create increased negative mood and decreased positive mood immediately after writing, compared with control writing. Postwriting minus prewriting change scores for each of the 5 moods were analyzed with 2 (group) x 3 (day) multivariate repeated-measure analyses of variance. There were significant group effects for all moods. As Table 2 shows, compared with the controls, the disclosure group experienced significant increases in sadness, fear, anger, and guilt and decreases in happiness from before to after writing. There were no group by day interactions, indicating that the mood effects of disclosure did not change over days. There was 1 day effect: disclosure writing led to increased guilt over days, which was paralleled by the control group, which showed decreased guilt on the first day followed by a return to baseline over days (F[2,44] = 3.18, p = .05).


View this table:
[in this window]
[in a new window]
  		TABLE 2. Change in Mood Ratings (Mean and SD) From Before to After Writing Across 3 Days for Disclosure and Control Groups (N = 47)a
 
Main Effects on Dependent Measures at 2-Month Follow-up
We tested for group main effects by first calculating change scores (follow-up minus baseline) for each of the health outcomes and then conducting analyses of variance on each change score. Table 1 presents the baseline, follow-up, and change scores for each group. There was only 1 significant effect. The disclosure group had a significant reduction in evaluative pain on the MPQ from baseline to follow-up, whereas the control group’s evaluative pain was unchanged (F[1,46] = 6.80, p = .01, {eta}2 = 0.13; note that covarying baseline evaluative pain and predicting the change score also indicated a significant group effect, p = .02.) A significantly higher percentage of disclosure participants (16 of 28, 57%) improved at least 1 point on the 0 to 5-point evaluative pain scale, compared with 4 of 20 controls (20%; {chi}2[N = 48] = 6.62, p = .01).

Moderators of the Effects of Disclosure Versus Control Writing
We tested the 3 potential moderators using regression analyses in which each outcome change score was predicted by entering the cross-product (group times moderator) into a hierarchical regression model after entering each main effect (group and moderator) in the first step. Predictors were first centered around 0 (ie, the sample mean was subtracted from each score so that the sample mean now equaled 0) to reduce possible problems with scaling and collinearity of the main effects and interaction terms, which is recommended when working with interaction terms that contain continuous variables (55). A significant interaction was taken as evidence of moderation by the baseline variable. To interpret interactions, a regression line predicting the outcome was plotted for each group as a function of the moderator, using sample values of ±1.0 SDs of the moderator. (Original rather than centered values were plotted to facilitate interpretation.)

Baseline ambivalence over emotional expression moderated group effects on changes in both daily disability (F[1,44] = 6.20, p = .017, {Delta}R2 = 0.12), and physical disability (F[1,44] = 5.11, p = .029, {Delta}R2 = 0.103). As Figure 1 shows, among disclosure participants, greater ambivalence predicted some reduction in daily disability (ß = -0.26, p = .18), but among controls, ambivalence predicted increased daily disability (ß = 0.43, p = .06). As shown in Figure 2, among disclosure participants, greater ambivalence predicted a slight reduction in physical disability (ß = -0.14, p = .47), but among controls, a substantial increase in physical disability (ß = 0.49, p = .03).



View larger version (12K):
[in this window]
[in a new window]
  		Figure 1. Baseline ambivalence over emotional expression moderates the effects of written emotional disclosure (filled circles) vs. positive control writing (open circles) on change in daily disability from baseline to 2-month follow-up (interaction p = .017).

 


View larger version (12K):
[in this window]
[in a new window]
  		Figure 2. Baseline ambivalence over emotional expression moderates the effects of written emotional disclosure (filled circles) vs. positive control writing (open circles) on change in physical disability from baseline to 2-month follow-up (interaction p = .029).

 
Catastrophizing also was a moderator of daily disability (F[1,44] = 5.19, p = .028, {Delta}R2 = 0.103). Similar to the moderation by ambivalence, greater catastrophizing predicted some improvement in daily disability among disclosure participants (ß = -0.24, p = .22) but worse disability among controls (ß = 0.40, p = .08).

Finally, baseline negative affect significantly moderated the group effect on changes in positive affect (F[1,44] = 4.75, p = .039, {Delta}R2 = 0.087). As shown in Figure 3, greater baseline negative affect predicted substantially improved positive affect among disclosure participants (ß = 0.51, p = .006) but was unrelated to the change in positive affect among controls (ß = 0.02, p = .95). Negative affect also moderated group effects on daily disability (F[1,44] = 4.75, p = .035, {Delta}R2 = 0.095). Like both ambivalence and catastrophizing, greater baseline negative affect predicted reduced daily disability among disclosure participants (ß = -0.37, p = .05) but a slight increase in daily disability among controls (ß = 0.23, p = .33).



View larger version (11K):
[in this window]
[in a new window]
  		Figure 3. Baseline negative affect moderates the effects of written emotional disclosure (filled circles) vs. positive control writing (open circles) on change in positive affect from baseline to 2-month follow-up (interaction p = .039).

 
Testing the Independence of Ambivalence and Catastrophizing From Negative Affect
Our final analyses provided more conservative tests of the significant moderating effects of ambivalence over emotional expression and catastrophizing on daily disability and physical disability, which are described above. These analyses tested whether ambivalence and catastrophizing moderated independently of negative affect. Note that ambivalence was marginally, positively correlated with negative affect (r[46] = 0.27, p = .07), and catastrophizing was significantly positively correlated with negative affect (r[46] = 0.45, p = .001), suggesting the potential for confounding with negative affect and the importance of testing the independent moderating effects of ambivalence and catastrophizing. (Ambivalence and catastrophizing were also positively correlated with each other; r[46] = 0.56, p < .001.)

We repeated the above significant hierarchical regressions that showed that ambivalence and catastrophizing moderated daily and physical disability, this time adding negative affect and the group by negative affect cross-product term to the model, before testing the group by ambivalence (or group by catastrophizing) term in a final step. In these models, baseline ambivalence remained a significant moderator of group effects predicting change daily disability (interaction term F[1,42] = 4.41, p = .042, {Delta}R2 = 0.084) and a marginally significant moderator predicting change in physical disability (F[1,42] = 3.46, p = .07, {Delta}R2 = 0.069). In contrast, baseline catastrophizing was no longer a significant moderator of change in daily disability (F[1,42] = 2.46, p = .12, {Delta}R2 = 0.049).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
CONCLUSION
 ACKNOWLEDGMENTS
 REFERENCES
 
This study demonstrated that writing about the stressful effects of CPP lead to lower levels of evaluative pain compared with writing about positive experiences. More important than this single main effect, however, is that several baseline variables predicted who benefited from disclosure rather than control writing. Women who were ambivalent about expressing their emotions, who engaged in catastrophizing, or who reported higher levels of negative affect before writing about stress showed the most improvement in disability or positive affect 2 months later, compared with women without these characteristics who wrote about stress, or with women with these characteristics who wrote about positive events.

Main Effects of Written Disclosure on Pain
Writing about the stressful consequences of CPP improved the evaluative dimension of pain; 57% of the disclosure women reported improvement, compared with 20% of controls. Yet this main effect should be interpreted cautiously. The groups differed on only 1 of several outcomes, and this construct was assessed with 1 item, suggesting limited reliability. Thus, consistent with a growing number of studies on disclosure among people with health problems (14–29), this study found only limited overall benefits of disclosure for women with CPP.

Individual Differences in Response to Written Disclosure
Individual differences, however, suggest that there are targeted benefits of writing about the stress of CPP. Being ambivalent about expressing emotion, engaging in pain catastrophizing, or having elevated negative affect before writing predicted improved health after disclosure relative to poorer health after positive writing. Women with higher ambivalence about expressing feelings showed reduced disability after disclosure writing, whereas there was a worsening of disability among ambivalent women who wrote about positive experiences. This finding suggests that ambivalent women benefit from the opportunity to disclose their stressful feelings in writing, presumably because writing provides a safe venue for disclosing negative feelings (31). Elevated catastrophizing also predicted a positive response to writing about the stress of CPP. This is consistent with the finding that catastrophizers benefit from writing about their feelings before a dental examination (35) and suggests that acknowledging and processing one’s negative emotions and helplessness in the face of pain is beneficial. Finally, higher baseline negative affect predicted improved positive affect and daily disability 2 months after disclosure writing, which agrees with studies suggesting that denying negative emotions predicts less benefit of disclosure (37, 38). Unlike other studies that did not find moderation by anxiety or depression in healthy samples (1, 36), this sample had elevated depression and had a wide range of scores on negative affect, which may have allowed moderator effects to occur.

Although ambivalence over emotional expression had a low positive correlation with negative affect, it was independent from negative affect in its moderation of outcomes. In contrast, catastrophizing was substantially correlated with negative affect (and with ambivalence), and negative affect accounted for the moderating effect of catastrophizing. Thus, both negative affect and ambivalence appear to have independent roles in predicting benefits of disclosure. Catastrophizing, however, does not predict independently from negative affect, and future research should examine whether catastrophizing contributes uniquely to health outcomes.

Types of Disclosure Writing and Control Writing
Participants adhered closely to the instructions for both conditions. Women assigned to write about the stress of CPP disclosed many private, difficult consequences, whereas control women wrote about a range of positive events. Furthermore, disclosure writers were more likely to have held back from disclosing these experiences to others, whereas the 2 groups rated their topics as equally meaningful. This suggests that the disclosure condition succeeded in eliciting previously inhibited stressors, and that the control condition was engaging and face valid.

Our study targeted stress secondary to CPP, whereas others have targeted general stress (11, 15). In this study, writing about secondary stress led to immediate increases in negative mood, just as general stress writing does. Furthermore, the secondary stressors disclosed cut across many life domains, including mood and functioning, sexuality, marital issues, sense of self and identity, and relationships. These observations raise the interesting hypothesis that targeting stressful consequences of a condition activates general stressful experiences in patients’ lives. Yet, writing about stress secondary to a condition is likely to be more acceptable to patients, especially those whose illnesses may be attributed to emotional factors. Research should directly compare general stress and secondary stress instructions with respect to their acceptability and their outcomes.

We used a positive control condition because we thought that it would have face validity as stress management technique and provide a better contrast to stress writing. Like another study that found no benefit of writing about positive events but improved working memory after writing about stress (5), we found no evidence that writing about positive events was helpful for the control group overall, or even among subgroups defined by baseline negative affect or ambivalence. Indeed, women with high ambivalence or catastrophizing who wrote about positive topics showed increased disability at follow-up. This finding suggests that there was a mismatch between these personality variables and the task of focusing on only positive events. Those women who were ambivalent or who had elevated catastrophizing and were assigned to the positive event group may have had to inhibit negative thoughts and feelings actively, which could lead to poorer health. This is consistent with an explanation and experimental evidence offered by Cole (56) as to why some of the early written disclosure studies found significant worsening of the control group rather than improvement of the experimental group. Adding an emotionally neutral writing condition or a no-writing control condition would provide a more conservative test of our hypotheses and help clarify whether the apparent worsening was caused by positive writing or was the natural health trajectory for these subgroups of women.

Recent studies suggest that some types of positive writing might be helpful. Writing about the benefits of difficult experiences (13, 57) or attempting to find meaning in trauma (9) can lead to health improvements that equal or surpass those stemming from writing solely about negative events. These latter approaches likely involve accessing and reflecting on negative experiences to create positive comparisons, and this integration of the positive with the negative may be a very useful tactic, as suggested by analyses of the language used in disclosure essays (58). Instructions to find positive meaning in one’s difficulties may be beneficial and also avoid the short-term mood impairment that accompanies writing only about stressors. In contrast, writing solely about positive experiences may not lead to health improvements, although the inclusion of a neutral writing condition is needed to test this directly.

Limitations and Caveats
This study has several limitations. First, although the participants met criteria for CPP, this syndrome has many etiologies, and this heterogeneity may obscure the success of an intervention. Although exploratory analyses (not shown) found no evidence that the presence or absence of an identified cause of the CPP moderated the effects of disclosure, it remains possible that people with different causes or types of pain will respond differently to disclosure. Second, the writing and follow-up assessments were completed at home rather than under supervision, and although these procedures were intended to decrease participant burden and increase external validity, they also probably reduced the effect sizes by increasing error variance. Also, although we obtained the participants’ writings, thus verifying completion, we could not verify whether the writing and ratings were conducted at the instructed times.

Third, attrition from the study was higher than anticipated, which raises concerns over the generalizability of the findings. Several factors likely contributed to the attrition. Some women dropped because of questionnaire burden or because they were bothered by the measures, particularly the illness-related items on the SIP. Some women from the disclosure group withdrew because writing about stressful consequences of CPP was upsetting, and some controls withdrew because that they were depressed at not having positive things to write about. Future studies should strive for larger samples, which will provide greater statistical power to examine both main and moderator effects and also will enhance generalizability.

Fourth, the majority of the sample had current or lifetime depression, which has several potential implications. We suspect that many of the women were seeking help, often in desperation, and some may have been disappointed that the study did not provide a more active intervention or even much interaction with the researcher. Depression may have led some women to have difficulty managing the negative affect aroused by the assessments or the disclosure writing task. Recall that writing-induced guilt increased over the 3 writing days in both groups, suggesting that writing repeatedly about either topic was somehow difficult for these women. This guilt effect, however, was not paralleled by similar effects for the other negative moods (eg, anger, sadness, fear), and we found no evidence that the increase in guilt predicted poorer 2-month outcomes (analyses not shown). Thus, the reliability and importance of this guilt finding are unclear. Nonetheless, the attrition, occasional negative reactions to writing, and writing-induced guilt effect suggest that this protocol was challenging to this sample, which may stem from their elevated levels of depression and distress. Also, it should be remembered that the findings should be generalized only to similar, primarily depressed samples of women with CPP.

Finally, researchers and clinicians should think more critically about the potency of written disclosure in general. Writing about stressful experiences for several days may benefit healthy young adults (1–9) and even people with medical problems who are either highly selected or generally emotionally stable (11–15). Yet it should not be surprising that patients with challenging health conditions, such as comorbid pain (especially of unknown origin) and depression, may not benefit from a total of 60 minutes of private writing. Psychological interventions that are much more intensive often have limited effectiveness with these populations. Thus, this study’s brief duration and intensity of disclosure writing (and positive writing) may have contributed to a lack of main effects. Studies may consider modifying writing procedures such as having more days of writing, giving guidance on how to structure the writing (10), providing feedback to the participants about their writing, or even or teaching coping strategies after writing.


   CONCLUSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSION
ACKNOWLEDGMENTS
 REFERENCES
 
In summary, women with CPP may derive some limited reduction in pain from writing about the stressful consequences of their condition. More importantly, however, mood and functioning appear to improve for women who are ambivalent about expressing their emotions or who have elevated negative affect. The limited overall clinical benefit of this approach, however, suggests its use not as a solitary intervention but as an adjunct to other medical and behavioral approaches to managing CPP. This study also reminds us that stress often results from a medical condition, and people interested in stress management may be able to target secondary stress rather than premorbid stressors, whose contribution to the current condition is often unclear. We believe that a focus on secondary stress will validate patients, particularly if their condition has an uncertain etiology, and will encourage greater participation in psychosocial interventions.


   ACKNOWLEDGMENTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSION
 ACKNOWLEDGMENTS
REFERENCES
 
The study was supported by dissertation grants from Sigma Xi and the Blue Cross Blue Shield of Michigan Foundation.

This article is based on the dissertation of the first author, conducted under the direction of the second author. The authors thank the Endometriosis Association for their assistance in recruitment and Lori Schram for her assistance in data collection. Data from this study were presented at the annual meeting of the American Psychosomatic Society, Monterey, CA, in 2001.

Received for publication June 26, 2003.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSION
 ACKNOWLEDGMENTS
 REFERENCES
 

  1. Pennebaker JW, Beall SK. Confronting a traumatic event: toward an understanding of inhibition and disease. J Abnorm Psychol 1986; 95: 274–81.[CrossRef][Medline]
  2. Cameron LD, Nicholls G. Expression of stressful experiences through writing: effects of a self-regulation manipulation for pessimists and optimists. Health Psychol 1998; 17: 84–92.[CrossRef][Medline]
  3. Esterling BA, Antoni MH, Fletcher MA, Margulies S, Schneiderman N. Emotional disclosure through writing or speaking modulates latent Epstein-Barr virus antibody titers. J Consult Clin Psychol 1994; 62: 130–40.[CrossRef][Medline]
  4. Greenberg MA, Wortman CB, Stone AA. Emotional expression and physical health: revising traumatic memories or fostering self-regulation? J Pers Soc Psychol 1996; 71: 588–602.[CrossRef][Medline]
  5. Klein K, Boals A. Expressive writing can increase working memory capacity. J Exp Psychol Gen 2001; 130: 520–33.[CrossRef][Medline]
  6. Lepore SJ, Greenberg MA. Mending broken hearts: effects of expressive writing on mood, cognitive processing, social adjustment, and health following a relationship breakup. Psychol Health 2002; 7: 547–60.[CrossRef]
  7. Pennebaker JW, Kiecolt-Glaser JK, Glaser R. Disclosure of traumas and immune function: health implications for psychotherapy. J Consult Clin Psychol 1988; 56: 239–45.[CrossRef][Medline]
  8. Petrie K, Booth RJ, Pennebaker JW, Davison KP, Thomas MG. Disclosure of trauma and immune response to a hepatitis B vaccination program. J Consult Clin Psychol 1995; 63: 787–92.[CrossRef][Medline]
  9. Ullrich PM, Lutgendorf SK. Journaling about stressful events: effects of cognitive processing and emotional expression. Ann Behav Med 2002; 24: 244–50.[CrossRef][Medline]
  10. Gidron Y, Duncan E, Lazar A, Biderman A, Tandeter H, Shvartzman P. Effects of guided written disclosure of stressful experiences on clinic visits and symptoms in frequent clinic attenders. Fam Pract 2002; 19: 161–6.[Abstract/Free Full Text]
  11. Smyth JM, Stone AA, Hurewitz A, Kaell A. Effects of writing about stressful experiences on symptom reduction in patients with asthma or rheumatoid arthritis. JAMA 1999; 281: 1304–9.[Abstract/Free Full Text]
  12. Solano L, Donati V, Pecci F, Persichetti S, Colaci A. Post-operative course after papilloma resection: effects of written disclosure of the experience in subjects with different alexithymia levels. Psychosom Med 2003; 65: 477–84.[Abstract/Free Full Text]
  13. Stanton AL, Danoff-Burg S, Sworowski LA, Collins CA, Branstetter AD, Rodriguez-Hanley A, Kirk SB, Austenfeld JL. Randomized, controlled trial of written emotional expression and benefit finding in breast cancer patients. J Clin Oncol 2002; 20: 4160–8.[Abstract/Free Full Text]
  14. de Moor C, Sterner J, Hall M, Warneke C, Gilani Z, Amato R, Cohen L. A pilot study of the effects of expressive writing on psychological and behavioral adjustment in patients enrolled in a phase II trial of vaccine therapy for metastatic renal cell carcinoma. Health Psychol 2002; 21: 615–9.[CrossRef][Medline]
  15. Kelley JE, Lumley MA, Leisen JCC. The health effects of emotional disclosure in rheumatoid arthritis. Health Psychol 1997; 16: 331–40.[CrossRef][Medline]
  16. Rosenberg HJ, Rosenberg SD, Ernstoff MS, Wolford GL, Amdur RJ, Elshamy MR, Bauer-Wu SM, Ahles TA, Pennebaker JW. Expressive disclosure and health outcomes in a prostate cancer population. Int J Psychiatry Med 2002; 32: 37–53.[CrossRef][Medline]
  17. Meyer T, Lumley M, Markowitz A, Macklem D, Leisen J, Lubetsky M, Lasichak L, Mosley A, Granda J. Written or verbal emotional disclosure about stress in rheumatoid arthritis. Psychosom Med 2003; 65: A-13.
  18. Mann T. Effects of future writing and optimism on health behaviors in HIV-infected women. Ann Behav Med 2001; 23: 26–33.[CrossRef][Medline]
  19. Richards JM, Beal WE, Seagal JD, Pennebaker JW. Effects of disclosure of traumatic events on illness behavior among psychiatric prison inmates. J Abnorm Psychol 2000; 109: 156–60.[CrossRef][Medline]
  20. Broderick J, Stone A, Smyth J, Kaell A. The feasibility and effectiveness of an expressive writing intervention for rheumatoid arthritis via home-based videotaped instructions. Ann Behav Med 2004; 27: 50–9.[CrossRef][Medline]
  21. Klapow JC, Schmidt SM, Taylor LA, Roller P, Li Q, Calhoun JW, Wallander J, Pennebaker J. Symptom management in older primary care patients: feasibility of an experimental, written self-disclosure protocol. Ann Int Med 2001; 134: 905–11.[Abstract/Free Full Text]
  22. Stroebe M, Stroebe W, Schut H, Zech E, van den Bout J. Does disclosure of emotions facilitate recovery from bereavement? Evidence from two prospective studies. J Consult Clin Psychol 2002; 70: 169–78.[CrossRef][Medline]
  23. Walker BL, Nail LM, Croyle RT. Does emotional expression make a difference in reactions to breast cancer? Oncol Nurs Forum 1999; 26: 1025–32.[Medline]
  24. Bower JE, Kemeny ME, Taylor SE, Fahey JL. Finding positive meaning and its association with natural killer cell cytotoxicity among participants in a bereavement-related disclosure intervention. Ann Behav Med 2003; 25: 146–55.[CrossRef][Medline]
  25. Batten SV, Follette VM, Rasmussen Hall ML, Palm KM. Physical and psychological effects of written disclosure among sexual abuse survivors. Behav Ther 2002; 33: 107–22.
  26. D’Souza P, Lumley M, Kraft C, Dooley J, Roberson T, Stanislawski B, Ramos M. Emotional disclosure and relaxation training for migraine and tension headaches: a randomized trial. Psychosom Med 2003; 65: A-6.
  27. Lumley M, Leegstra S, Provenzano K, Warren V. The health effects of writing about stress on physically symptomatic young adults. Psychosom Med 1999; 61: 89.
  28. Tolks T. The influence of written emotional disclosure on conception rates in patients undergoing in-vitro fertilization. MSc thesis. University of Auckland, New Zealand; 1999.
  29. Gidron Y, Peri T, Connolly JF, Shalev AY. Written disclosure in posttraumatic stress disorder: is it beneficial for the patient? J Nerv Ment Dis 1996; 184: 505–7.[CrossRef][Medline]
  30. Lumley MA, Tojek TM, Macklem DJ. The effects of written and verbal disclosure among repressive and alexithymic people. In: Lepore SJ, Smyth JM, editors. The writing cure: how expressive writing promotes health and emotional well-being. Washington, DC: American Psychological Association; 2002. p. 75–95.
  31. King LA, Emmons RA. Conflict over emotional expression: psychological and physical correlates. J Pers Soc Psychol 1990; 58: 864–77.[CrossRef][Medline]
  32. Smyth J, Anderson C, Hockemeyer J, Stone A. Does emotional non-expressiveness or avoidance interfere with writing about stressful life events? An analysis in patients with chronic illness. Psychol Health 2002; 17: 561–9.[CrossRef]
  33. Paez D, Velasco C, Gonzalez JL. Expressive writing and the role of alexithymia as a dispositional deficit in self-disclosure and psychological health. J Pers Soc Psychol 1999; 77: 630–41.[CrossRef][Medline]
  34. Rosenstiel AK, Keefe FJ. The use of coping strategies in chronic low back pain patients: relationship to patient characteristics and current adjustment. Pain 1983; 17: 33–44.[CrossRef][Medline]
  35. Sullivan MJL, Neish N. The effects of disclosure on pain during dental hygiene treatment: the moderating role of catastrophizing. Pain 1999; 79: 155–63.[CrossRef][Medline]
  36. Pennebaker JW, Colder M, Sharp LK. Accelerating the coping process. J Pers Soc Psychol 1990; 58: 523–37.
  37. Esterling BA, Antoni MH, Kumar H, Schneiderman N. Emotional repression, stress disclosure responses, and Epstein-Barr viral capsid antigen titers. Psychosom Med 1990; 52: 397–410.[Abstract/Free Full Text]
  38. Habbal R. Emotional disclosure in women: the moderating effects of personality. Unpublished master’s thesis. San Diego State University; 1999.
  39. Watson D, Clark LA. Negative affectivity: the disposition to experience aversive emotional states. Psychol Bull 1984; 96: 465–90.[CrossRef][Medline]
  40. Steege JF, Stout AL, Somkuti SG. Chronic pelvic pain in women: toward an integrative model. Obstet Gynecol Surv 1993; 48: 95–110.[Medline]
  41. Mathias SD, Kupperman M, Liberman RF, Lipschutz RC, Steege JF. Chronic pelvic pain: prevalence, health-related quality of life, and economic correlates. Obstet Gynecol 1996; 87: 321–7.[Abstract]
  42. Baker PK. Muskuloskeletal origins of chronic pelvic pain. Obstet Gynecol Clin North Am 1993; 20: 719–42.[Medline]
  43. Rapkin AJ, Mayer EA. Gastroenterologic causes of chronic pelvic pain. Obstet Gynecol Clin North Am 1993; 20: 663–83.[Medline]
  44. Summit RL. Urogynecologic causes of chronic pelvic pain. Obstet Gynecol Clin North Am 1993; 20: 685–99.[Medline]
  45. Hodgkiss AD, Watson JP. Psychiatric morbidity and illness behavior in women with chronic pelvic pain. J Psychosom Res 1993; 38: 3–9.
  46. Kellner R, Slocumb JC, Rosenfeld RC, Pathak D. Fears and beliefs in patients with the pelvic pain syndrome. J Psychosom Res 1988; 32: 303–10.[CrossRef][Medline]
  47. Grace VM. Problems of communication, diagnosis, and treatment experienced by women using the New Zealand health services for chronic pelvic pain: a quantitative analysis. Health Care Women Int 1995; 16: 521–35.[Medline]
  48. Smyth JM. Written emotional expression: effect sizes, outcome types, and moderating variables. J Consult Clin Psychol 1998; 66: 174–84.[CrossRef][Medline]
  49. Watson D, Clark LA. The PANAS-X: manual for the Positive and Negative Affect Schedule-Expanded Form. University of Iowa; 1994.
  50. Melzack R, Katz J. The McGill Pain Questionnaire: appraisal and current status. In: Turk DC, Melzack R, editors. Handbook of pain assessment. New York: Guilford; 1992. p. 152–68.
  51. Turk DC, Rudy TE, Salovey P. The McGill Pain Questionnaire reconsidered: confirming the factor structures and examining appropriate uses. Pain 1985; 21: 385–97.[CrossRef][Medline]
  52. Bergner M, Bobbitt RA, Carter WB, Gilson BS. The Sickness Impact Profile: development and final revision of a health status measure. Med Care 1981; 19: 787–805.[Medline]
  53. Follick MJ, Smith TW, Ahern DK. The Sickness Impact Profile: a global measure of disability in chronic low back pain. Pain 1985; 21: 67–76.[CrossRef][Medline]
  54. Keefe FJ, Lumley MA, Anderson T, Lynch T, Studts J, Carson K. Pain and emotion: new research directions. J Clin Psychol 2001; 57: 587–607.[CrossRef][Medline]
  55. Aiken LS, West, SG. Multiple regression: testing and interpreting interactions. Newbury Park, CA: Sage; 1991.
  56. Cole SW. Negative physical health effects of psychological inhibition in disclosure study control groups. Unpublished manuscript.
  57. King LA, Miner KN. Writing about the perceived benefits of traumatic events: implications for physical health. Pers Soc Psychol Bull 2000; 26: 220–30.[Abstract/Free Full Text]
  58. Pennebaker JW, Mayne TJ, Francis ME. Linguistic predictors of adaptive bereavement. J Pers Soc Psychol 1997; 72: 863–71.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 The OncologistHome page
N. P. Morgan, K. D. Graves, E. A. Poggi, and B. D. Cheson
Implementing an expressive writing study in a cancer clinic.
Oncologist, February 1, 2008; 13(2): 196 - 204.
[Abstract] [Full Text] [PDF]

Home page
 Journal of Family NursingHome page
E. Duncan, Y. Gidron, E. Rabin, L. Gouchberg, A. M. Moser, and J. Kapelushnik
The Effects of Guided Written Disclosure on Psychological Symptoms Among Parents of Children With Cancer
Journal of Family Nursing, August 1, 2007; 13(3): 370 - 384.
[Abstract] [PDF]

Home page
 GerontologistHome page
C. S. Mackenzie, U. J. Wiprzycka, L. Hasher, and D. Goldstein
Does Expressive Writing Reduce Stress and Improve Health for Family Caregivers of Older Adults?
Gerontologist, June 1, 2007; 47(3): 296 - 306.
[Abstract] [Full Text] [PDF]

Home page
 Pers Soc Psychol BullHome page
J. D. Creswell, S. Lam, A. L. Stanton, S. E. Taylor, J. E. Bower, and D. K. Sherman
Does Self-Affirmation, Cognitive Processing, or Discovery of Meaning Explain Cancer-Related Health Benefits of Expressive Writing?
Pers Soc Psychol Bull, February 1, 2007; 33(2): 238 - 250.
[Abstract] [PDF]

Home page
 J Pediatr PsycholHome page
L. J. Warner, M. A. Lumley, R. J. Casey, W. Pierantoni, R. Salazar, E. M. Zoratti, R. Enberg, and M. R. Simon
Health Effects of Written Emotional Disclosure in Adolescents with Asthma: A Randomized, Controlled Trial
J. Pediatr. Psychol., July 1, 2006; 31(6): 557 - 568.
[Abstract] [Full Text] [PDF]

Home page
 Adv. Psychiatr. Treat.Home page
K. A. Baikie and K. Wilhelm
Emotional and physical health benefits of expressive writing
Advan. Psychiatr. Treat., September 1, 2005; 11(5): 338 - 346.
[Abstract] [Full Text] [PDF]

Home page
 Psychosom. Med.Home page
J. E. Broderick, D. U. Junghaenel, and J. E. Schwartz
Written Emotional Expression Produces Health Benefits in Fibromyalgia Patients
Psychosom Med, March 1, 2005; 67(2): 326 - 334.
[Abstract] [Full Text] [PDF]